One thing that you guys taught me is how VIX futures (and options) at far out months are a lot less sensitive to spot VIX changes than the early months. This makes sense given that VIX is mean reverting, and the futures are European style (maybe just the former does the trick and the latter is not really important). But anyways. So the spot/early month VIX spiking more than the later months makes complete sense. But is there ever a situation where the VIX later month contracts ever spike more than than the spot VIX? I can't think of a scenario, although I'm probably missing something. Thanks!
Anything is possible... However, it's difficult to imagine circumstances under which far VIX futures jump lots more than the fronts. Ergo, it's highly unlikely. Also, for the record, futures can't be European or American...
Sure it happens and actually it happens more often than you would think intuitively. It is certainly not impossible or even highly unlikely. Off the top of my head i usually see it happen during curve steepening regimes(contango) or flattening(backwardation) regardless of vix direction and the switch back and forth. In fact, you can even have the fronts(M1 and M2) fall while mid-back curve rises. It can also happen due to event vol repricing or when a large player is rolling inventory. Additionally, you can also have the front part of the curve fall less than the back end so sensitivity to spot is certainly not constant in either direction. There are probably a few other scenarios that escape me for the moment and even though it is not a daily or prevalent occurrence it would be best not to position yourself as if it would never happen.
Well, in fairness, when it happens it can hardly be called a “spike”, it's usually relatively mild transformation in the shape of the curve. Usually happens when the curve is in a flux state and vol component gets re-priced while the S&P index itself is relatively quiet. The closest I’ve seen to a spike (i.e. when you get a significant relative move) would event pricing, for example the "fiscal cliff" type situations. Since most of the VIX and VIX futures variance is due to the movement of the S&P index (about 70%), the term structure of skew has a dominant effect on the relative movement. The rest of the movement can come from the actual changes in the S&P volatility surface (well, more like vix futures driving those changes, but that's beside the point). Also, because of the forward var arbitrage relationship, it's pretty hard (read "almost impossible") to have a massive spike in VIX futures without moving the index, even if the move is initiated in the futures market like it did on the Feb 5th.
Perhaps we should actually define a spike In all seriousness though, I didn't realize the OP asked about a spike. My brain must have somehow skipped over that word. So if we are talking about a difference of handle or more in the delta of M1 and let's say M6, I would agree. I don't think I have seen that since VX inception and whenever I've seen it come close it's not exactly in the sense the OP is asking. However, I often do see 30 to 40 cent kinks and when we are talking about calendar spreads clocking in at $1,000 per tick I wouldn't exactly call that mild.
I was gonna google that for you, but the first thing that comes up is an Israeli anti-tank missile: https://en.wikipedia.org/wiki/Spike_(missile) Didn't we see like a 1 point dislocation between the index and the front contract going into the congress meeting in 2012? I don't have access to the data, so can't look it up. Quarter to a half vol is what I'd say is a normal "pure vol move". You can have a fair amount of idiosyncratic movement there that does not really violate any arb relationships and does not really cause any panic. PS. In any case, I was not trying to start an argument
No arguement here and yes, the fiscal cliff does ring some unpleasant bells in my mind as well as brexit or perhaps it was frexit. I really try to forget.
Thanks everyone! Yea, I meant an actual spike. So, I'm not talking about a situation where the VIX spot is at 17, future is at 17.50, spot drops to 12, future only drops to 16. That seems like it would happen all the time given the relative insensitivity of the out months. I was just trying to think through a time when the spot would stay relatively unchanged but the forward months would spike. VIX today is at 12, out month is at 12, all of a sudden out month goes to 17 but spot only goes to 13, for example. Would just seem weird for the market to predict price uncertainty in the future - if the market sees an event that is coming up that will create uncertainty seems it would be priced into the S&P options that make up the VIX right away. Thanks!